DESCRIPTION (Taken from the Application): Demonstration of the power of structural genomics using it to address a significant problem in human health, the pathogenicity of Mycobacterium tuberculosis, is proposed. The structures of over 400 functionally important proteins from M. tuberculosis will be determined and analyzed. The database of structural and functional information to be constructed will form a basis for understanding M. tuberculosis pathogenesis and will enable structure-based design of novel drugs. The structures will include about 40 novel folds and 200 new structural families, providing a foundation for structure prediction of related proteins. To accomplish this, an approach has been developed that will deliver structures and analyses of high value in a cost-effective manner. Functionally important proteins will be targeted through the use of genetic screens and genome-scale functional assignments. The green-fluorescent, protein-based screening system will be used to optimize proteins for expression, solubility and methionine content. Crystallization will be carried out by a low-cost system that combines automation of their stochastic screening protocol with image analysis of droplets. Selenomethionine MAD X-ray data collection on characterized crystals at synchrotron facilities will be emphasized, with concurrent structure solution using automated software. Synchrotron time sufficient for collection of 300 MAD structures per year has been secured for this program. Structural data will be systematically analyzed for fold assignment, similarity to other proteins, and local motifs. Analysis of structures determined will use new function prediction methods that will guide biochemical tests of function. A consortium approach to structural genomics will be implemented that allows a world-wide effort to be focused on a defined set of targets. The Consortium for Structural Genomics consists of 24 laboratories from 13 institutions in 6 countries. Consortium laboratories are collectively responsible for 3.3 percent of all protein structures in the Protein Data Bank and have extensive records of methods development. Consortium members have carried out a pilot project on the structural genomics of a hyperthermophile that has identified bottlenecks and resulted in development of methodologies for high-throughput structure determination and analysis. The Consortium will have centralized facilities that will carry out an increasing fraction of routine tasks such as protein production, crystallization and X-ray data collection. The structural and functional information obtained in this project is to be placed in the public domain by timely deposition in publicly available databases.